of simply supported reinforced concrete beam
TRANSCRIPT
sensors
Communication
Influence of Temperature on the Natural Vibration Characteristicsof Simply Supported Reinforced Concrete Beam
Yanxia Cai 1,2,3, Kai Zhang 4, Zhoujing Ye 5 , Chang Liu 5, Kaiji Lu 1,2,3 and Linbing Wang 5,6,*
�����������������
Citation: Cai, Y.; Zhang, K.; Ye, Z.;
Liu, C.; Lu, K.; Wang, L. Influence of
Temperature on the Natural Vibration
Characteristics of Simply Supported
Reinforced Concrete Beam. Sensors
2021, 21, 4242. https://doi.org/
10.3390/s21124242
Academic Editor: Amir H. Alavi
Received: 26 May 2021
Accepted: 17 June 2021
Published: 21 June 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
1 Beijing Zhonglu Gaoke Highway Technology Co., Ltd., Beijing 100088, China; [email protected] (Y.C.);[email protected] (K.L.)
2 Research and Development Center of Transport Industry of New Materials, Technologies Application forHighway Construction and Maintenance, Beijing 100088, China
3 Research Institute of Highway Ministry of Transport, Beijing 100088, China4 China Construction Third Bureau First Engineering Co., Ltd., Wuhan 430040, China; [email protected] National Center for Materials Service Safety, University of Science and Technology Beijing, Beijing 100083,
China; [email protected] (Z.Y.); [email protected] (C.L.)6 Virginia Tech, Blacksburg, VA 24061, USA* Correspondence: [email protected]
Abstract: Natural vibration characteristics serve as one of the crucial references for bridge monitoring.However, temperature-induced changes in the natural vibration characteristics of bridge structuresmay exceed the impact of structural damage, thus causing some interference in damage identification.This study analyzed the influence of temperature on the natural vibration characteristics of simplysupported beams, which is the most widely used bridge structure. The theoretical formula for thevariation of the natural frequency of simply supported beams with temperature was proposed. Theelastic modulus of simply supported beams in the range of−40 ◦C to 60 ◦C was acquired by means ofthe falling ball test and the theoretical formula and was compared with the elastic modulus obtainedby the three-point bending test at room temperature (20 ◦C). In addition, the Midas/Civil finite-element simulation was carried out for the natural frequency of simply supported beams at differenttemperatures. The results showed that temperature was the main factor causing the variation of thenatural frequency of simply supported beams. The linear negative correlation between the naturalfrequency of simply supported beams and their temperature were observed. The natural frequencyof simply supported beams decreased by 0.148% for every 1 ◦C increase. This research contributedto the further understanding of the natural vibration characteristics of simply supported beamsunder the influence of temperature so as to provide references for natural frequency monitoring anddamage identification of beam bridges.
Keywords: temperature; natural vibration characteristics; simply supported beam; beam bridge;Midas/Civil
1. Introduction
The natural vibration characteristics of bridges, including frequency, vibration mode,and damping, are affected by structural stiffness and the extent of damage, which can pro-vide a reference basis for bridge design and comprehensive performance evaluation [1–4].However, environmental factors, such as temperature, humidity, wind speed, and ex-tremely harsh environments, can also cause significant changes in the natural vibrationcharacteristics of the structure, which may even be greater than the changes in the natu-ral vibration characteristics caused by actual damage, leading to difficulties in damageidentification technologies for bridge monitoring [5–9].
As an important reference basis in damage identification, natural vibration charac-teristics play an essential role in bridge monitoring. In recent years, many researchershave explored the evolutionary law of natural bridge frequency and environmental factors.
Sensors 2021, 21, 4242. https://doi.org/10.3390/s21124242 https://www.mdpi.com/journal/sensors
Sensors 2021, 21, 4242 2 of 14
Farrar CR et al. [10] carried out a study on the modal parameters of beam bridges based onenvironmental factors and found that the natural frequency of beam bridges varied by 5%over 24 h. Therefore, it can be concluded that the bridge modal change is mainly related tothe bridge temperature. Additionally, Liu and DeWolf [11] conducted a one-year observa-tion on a curved concrete box girder bridge to investigate the dynamic characteristics. Theresults showed that temperature change had an impact on the bridge’s modal frequency,with a maximum of a 6% change in the bridge’s modal frequency when the temperaturechanged to 39 ◦C over a year. Through regular monitoring of the temperature and modalfrequency of the reinforced concrete beam model, Liu et al. [5] found that, when the tem-perature experienced a change from −14.1 ◦C to 22.8 ◦C, the first four modal frequencieschanged by 5–12%. This variation is sufficient to cover the variation in modal frequencydue to structural damage. Sun Limin et al. [12] conducted research on the relationshipbetween temperature, humidity, and bridge modal frequency based on the concrete contin-uous beam bridge model in the laboratory. The experimental results suggested that thenatural frequency of the bridge decreased linearly with the increase of temperature. Whenthe boundary condition remained constant, the influence of air humidity on the structurefrequency was less than that of temperature. In addition, changes in temperature may alterthe boundary conditions of the bridge and thus affect the structural frequency. The abovestudies proved that the temperature has a significant influence on the bridge’s naturalfrequency among the environmental factors. Therefore, it is necessary to investigate theeffect of temperature on the natural vibration characteristics of bridges.
In order to study the influence of temperature on the natural vibration characteristicsof bridges, the use of field monitoring of the bridge, or the finite-element method, is themain strategy to obtain the relationship between bridge natural frequency and temperature.However, the influence of other factors cannot be excluded in field monitoring of thebridge, resulting in different conclusions. Most of the studies [5,12–18] indicate that thefrequency decreases with the increase of temperature, though there are also papers [19–21]that hold the opposite views. It should be noted that the long-term monitoring data cannotexclude the influence of material property changes and structural damage on naturalvibration characteristics. Changes in the natural vibration characteristics of bridges arenot necessarily caused by temperature. Changes in the self-vibration characteristics ofbridges are also not necessarily caused by temperature, as there are many environmentalfactors simultaneously in effect. Furthermore, the influence of temperature on the dynamiccharacteristics of the bridge is different for various bridge types and boundary conditions,which is the main reason why some scholars reached different conclusions.
The reinforced concrete simply supported beam bridge is the earliest and most widelyused bridge structure, which consists of beams supported by movable bearings and articu-lated bearings at both ends (namely: simply supported beam) as the main load-bearingstructure. The factors that influence the natural frequency of simply supported beambridges due to temperature changes are as follows [22,23]: (1) The change of temperaturewill cause the change in elastic modulus of structural materials (such as concrete and thereinforcing bar); (2) The change of temperature will cause changes in structure size; (3)In the statically indeterminate structure, the change of temperature can cause secondarystress; and (4) The change of temperature will cause changes in boundary conditions ofthe structure, which are mainly manifested by changes in abutment stiffness. Yang [24]and Li [25] both analyzed the influence of ambient temperature on the natural vibrationcharacteristics of concrete beams through laboratory experiments, and proposed that thereis a negative correlation between the bridge’s natural frequency and the ambient temper-ature. However, both experiments were performed at ambient temperature, which, inaddition to the temperature of the simply supported beam have relatively small changeswith the temperature control range above 0 ◦C. It is necessary to conduct experiments overa wider range of temperature variations so as to ensure the universality and reliability ofthe conclusions.
Sensors 2021, 21, 4242 3 of 14
In this paper, the influence of temperature on the natural vibration characteristicsof reinforced concrete simply supported beam was analyzed in the laboratory. The theo-retical calculation formula for simply supported beam natural frequency variation withtemperature was presented. The variation law of the natural frequency of testing beamswithin the temperature range of −40 ◦C to 60 ◦C was explored by means of the falling balltest and three-point bending test. The Midas/Civil finite-element simulation method wasalso applied to analyze the natural frequency of the testing beam at different temperatures.Through theoretical analysis, experimental analysis, and simulation analysis, the relation-ship between the natural frequency and the temperature was derived, which can providereferences for actual bridge natural frequency monitoring and damage identification.
2. Theoretical Analysis on the Influence of Temperature on the Natural Frequency ofSimply Supported Beam
As the simply supported beam is a statically determinate structure without abutmentstiffness, temperature will not cause changes in secondary structural stress and bound-ary conditions. The influence of temperature on elastic modulus and structure size isanalyzed below.
2.1. Influence of Temperature on Elastic Modulus
The influence of temperature on a concrete elastic modulus far exceeds that of areinforced elastic modulus. Concrete is the main material of the simply supported beam.Therefore, the influence of temperature on reinforcing bar was ignored. The analysis ofthe influence of temperature on the elastic modulus of the simply supported beam was inaccordance with the calculation formula in the European Concrete Specification CEB-FIPModel Code 2010 [26]:
ET = E20◦C[1− θE(T− 20)] (1)
where,
ET: Elastic modulus (Mpa) of concrete at T temperature;T: Temperature of concrete (◦C);E20◦C: Elastic modulus of concrete (MPa) at 20 ◦C;θE: Temperature coefficient of elastic modulus, with a value of 0.003.
2.2. Influence of Temperature on Structure Size
The influence of temperature on the structure size of the simply supported beamis mainly manifested in the elongation or shortening of the beam and the change of thesection inertia moment, which is related to the properties of the material. For the rectangularequal section simply supported beam, the change in beam length can be obtained by thefollowing formula:
ll = (1 + α∆T)l (2)
where,
l1: Beam length after temperature change;α: Linear expansion coefficient with reinforcement of 1.2× 10−6 and concrete of 1.0 × 10−5;∆T : Temperature change;l: Initial beam length;
The variation of beam width b and beam height h can be expressed as:
bl = (1 + vα∆T)b (3)
hl = (1 + vα∆T)h (4)
Sensors 2021, 21, 4242 4 of 14
where, v represents Poisson’s ratio, with concrete Poisson’s ratio v = 0.2 and reinforcedPoisson’s ratio v = 0.31. On the basis of section inertia moment I = bh3
12 , the variation ofthe section inertia moment can be expressed as:
I =(1 + vα∆T)4bh3
12(5)
2.3. Influence of Temperature on the Natural Frequency of Simply Supported Beam
According to the derivation method in the literature [2,17,23], the relationship betweenthe natural frequency and the temperature was explored using the first-grade rectangularsection simply supported beam as a model. Let the length, width, and height be l, b, and hrespectively, as shown in Figure 1.
Sensors 2021, 21, x FOR PEER REVIEW 4 of 15
bl = (1 + vα∆T)b (3)
hl = (1 + vα∆T)h (4)
where, v represents Poisson’s ratio, with concrete Poisson’s ratio v = 0.2 and reinforced
Poisson’s ratio v = 0.31. On the basis of section inertia moment I =bh3
12, the variation of
the section inertia moment can be expressed as:
I =(1 + vα∆T)4bh3
12 (5)
2.3. Influence of Temperature on the Natural Frequency of Simply Supported Beam
According to the derivation method in the literature [2,17,23], the relationship be-
tween the natural frequency and the temperature was explored using the first-grade rec-
tangular section simply supported beam as a model. Let the length, width, and height be
l, b, and h respectively, as shown in Figure 1.
Figure 1. Diagram of the simply supported beam.
The basic transverse vibration equation for the undamped homogeneous beam is as
follows [2,17]:
∂4v(x, t)
∂x4+
ρA
EI×
∂2v(x, t)
∂t2= 0 (6)
The formula for the free vibration frequency of the model is:
fn =n2π
2l2√
EI
ρA (7)
where,
fn: nth order modal frequency
l: Calculated span lengths of beam
E: Young’s modulus
I: Second moment of area
ρ: Mass density
A: Sectional area
Standard vibration mode function:
φn(x) = sinnπx
l (8)
The fundamental frequency temperature is 20 °C. Function (1) is adopted, consider-
ing the change of the material elastic modulus with temperature:
ET = E(1 − θE∆T) (9)
By substituting Formulas (2), (5), and (9) into Formula (7), the natural frequency of
the simply supported beam considering the effect of temperature is:
Figure 1. Diagram of the simply supported beam.
The basic transverse vibration equation for the undamped homogeneous beam is asfollows [2,17]:
∂4v(x, t)∂x4 +
ρAEI× ∂2v(x, t)
∂t2 = 0 (6)
The formula for the free vibration frequency of the model is:
fn =n2π
2l2
√EIρA
(7)
where,
fn: nth order modal frequencyl: Calculated span lengths of beamE: Young’s modulusI: Second moment of areaρ: Mass densityA: Sectional area
Standard vibration mode function:
ϕn(x) = sinnπx
l(8)
The fundamental frequency temperature is 20 ◦C. Function (1) is adopted, consideringthe change of the material elastic modulus with temperature:
ET = E(1− θE∆T) (9)
By substituting Formulas (2), (5), and (9) into Formula (7), the natural frequency ofthe simply supported beam considering the effect of temperature is:
fTn =
n2π
2l2
√EIρA·√(1− θE∆T)(1 + vα∆T)2
(1 + α∆T)2 = λIλlλEn2π
2l2
√EIρA
(10)
where,
λI: Influence coefficient of inertia moment, λI = (1 + vα∆T)2
Sensors 2021, 21, 4242 5 of 14
λl: Influence coefficient of beam length, λl =1
(1+α∆T)2
λE: Influence coefficient of elastic modulus, λE =√
1− θE∆T
With model beam α = 1.0× 10−5, v = 0.2, and θE = 0.003 into Function (10), thechange rate of the natural frequency of the simply supported reinforced concrete beam is:
∆fn =
(fTn
fn− 1
)× 100% =
√(1− 0.003∆T)
(1 + 0.2× 10−5∆T
)2
(1 + 1.0× 10−5∆T)2 − 1 (11)
Taking 0–50 ◦C with the interval of 2 ◦C and assigning the value to ∆T in Formula (11),the corresponding value of ∆fn can be obtained. The curve fitting is shown in Figure 2.
Sensors 2021, 21, x FOR PEER REVIEW 5 of 15
fnT =
n2π
2l2√
EI
ρA∙
√(1 − θE∆T)(1 + vα∆T)2
(1 + α∆T)2= λIλlλE
n2π
2l2√
EI
ρA (10)
where,
λI: Influence coefficient of inertia moment, λI = (1 + vαΔT)2
λl: Influence coefficient of beam length, λl =1
(1+αΔT)2
λE: Influence coefficient of elastic modulus, λE = √1 − θEΔT
With model beam α = 1.0 × 10−5 , v= 0.2, and θE = 0.003 into Function (10), the
change rate of the natural frequency of the simply supported reinforced concrete beam is:
Δfn = (fn
T
fn− 1) × 100% =
√(1 − 0.003∆T)(1 + 0.2 × 10−5∆T)2
(1 + 1.0 × 10−5∆T)2− 1 (11)
Taking 0–50 °C with the interval of 2 °C and assigning the value to ∆T in Formula
(11), the corresponding value of ∆fn can be obtained. The curve fitting is shown in Figure
2.
Figure 2. The relationship between ∆𝑓𝑛 and ∆𝑇.
It can be seen from Figure 2 that R2 = 0.9999, which indicates that the fitting curve
error was small, and ∆fn = −0.16%, which means that the frequency change of the model
beam caused by the change of material characteristics and geometric size was −0.16%
when the temperature rose by 1 °C. Taking 20 °C as the fundamental frequency tempera-
ture, the influence coefficients of each grade of natural frequency in the ultimate temper-
ature drop of 60 °C were obtained as follows:
Influence coefficient of elastic modulus: λE = 1.08628
Influence coefficient of inertia moment: λI = 0.99976
Influence coefficient of beam length: λl = 1.00120
Considering the ultimate cooling ∆𝑇 = 60 °C, the elastic modulus of the simply sup-
ported beam causes the frequency to increase by 8.628%; the section inertia moment
causes the frequency to decrease by 0.024%; the beam length causes the frequency to in-
crease by 0.12%; and the comprehensive effect of structure size causes the frequency to
increase by 0.096%. In comparison, the changes in section inertia moment and beam length
are negligible. The change in the natural frequency of the simply supported beam with
temperature is mainly related to the material elastic modulus, while the change in the
structure size has almost no effect on the frequency.
Figure 2. The relationship between ∆fn and ∆T.
It can be seen from Figure 2 that R2 = 0.9999, which indicates that the fitting curve errorwas small, and ∆fn = −0.16%, which means that the frequency change of the model beamcaused by the change of material characteristics and geometric size was −0.16% when thetemperature rose by 1 ◦C. Taking 20 ◦C as the fundamental frequency temperature, theinfluence coefficients of each grade of natural frequency in the ultimate temperature dropof 60 ◦C were obtained as follows:
Influence coefficient of elastic modulus: λE = 1.08628Influence coefficient of inertia moment: λI = 0.99976Influence coefficient of beam length: λl = 1.00120
Considering the ultimate cooling ∆T = 60 ◦C, the elastic modulus of the simplysupported beam causes the frequency to increase by 8.628%; the section inertia momentcauses the frequency to decrease by 0.024%; the beam length causes the frequency toincrease by 0.12%; and the comprehensive effect of structure size causes the frequencyto increase by 0.096%. In comparison, the changes in section inertia moment and beamlength are negligible. The change in the natural frequency of the simply supported beamwith temperature is mainly related to the material elastic modulus, while the change in thestructure size has almost no effect on the frequency.
3. Experimental Study on the Influence of Temperature on the Natural Frequency ofSimply Supported Beam3.1. Test Scheme
The test scheme is shown in Figure 3, and was mainly divided into three steps.
Sensors 2021, 21, 4242 6 of 14
Sensors 2021, 21, x FOR PEER REVIEW 6 of 15
3. Experimental Study on the Influence of Temperature on the Natural Frequency of
Simply Supported Beam
3.1. Test Scheme
The test scheme is shown in Figure 3, and was mainly divided into three steps.
Figure 3. Test scheme.
Step 1: Three reinforced concrete testing beams made of C30 concrete and HRB400
reinforcing bar were prefabricated, with lengths of 800 mm, widths of 80 mm, and heights
of 100 mm.
Step 2: The temperature of testing beams was controlled. The No. 1 and No. 2 testing
beams were placed into the temperature box as parallel test models, and the No. 3 testing
beam was reserved. The model of the temperature box was water+bai ag ETC 550-1, with
the working temperature of −150 °C to 600 °C. The temperature in the temperature box
was adjusted to 20 °C. To ensure that the internal temperature of the testing beam was
consistent with the external temperature, it was kept at a constant temperature for 3 h
after the temperature box reached 20 °C. In order to make the test results more accurate,
the test was carried out immediately after the No. 1 testing beam was taken out, while the
No. 2 testing beam was insulated in the temperature box.
Step 3: First, the testing beam was fixed. At 100 mm from the left and right ends of
the testing beam, steel bars with a diameter of 20 mm were respectively used as a fixed
end abutment and a sliding end abutment. Before the testing, the upper end of the testing
Figure 3. Test scheme.
Step 1: Three reinforced concrete testing beams made of C30 concrete and HRB400reinforcing bar were prefabricated, with lengths of 800 mm, widths of 80 mm, and heightsof 100 mm.
Step 2: The temperature of testing beams was controlled. The No. 1 and No. 2 testingbeams were placed into the temperature box as parallel test models, and the No. 3 testingbeam was reserved. The model of the temperature box was water+bai ag ETC 550-1, withthe working temperature of −150 ◦C to 600 ◦C. The temperature in the temperature boxwas adjusted to 20 ◦C. To ensure that the internal temperature of the testing beam wasconsistent with the external temperature, it was kept at a constant temperature for 3 h afterthe temperature box reached 20 ◦C. In order to make the test results more accurate, the testwas carried out immediately after the No. 1 testing beam was taken out, while the No. 2testing beam was insulated in the temperature box.
Step 3: First, the testing beam was fixed. At 100 mm from the left and right ends ofthe testing beam, steel bars with a diameter of 20 mm were respectively used as a fixedend abutment and a sliding end abutment. Before the testing, the upper end of the testingbeam was divided into four equal lengths. The center position of the three bisectors wasmarked as the landing point of small balls. Then, the acceleration sensor (Type: KT1100L,Measurement range:5 g) was fixed with silicone rubber at a distance of 300 mm from oneend of the testing beam. The oscilloscope (Type: RIGOL MSO1104) was adjusted to the“SINGLE” state. By controlling the air compressors (Type: AFS-550), the ball was fixed
Sensors 2021, 21, 4242 7 of 14
by the fixture. The ball fell freely onto the testing beam. After landing, the height wasadjusted until the oscilloscope showed a smooth fluctuation curve. By controlling the aircompressors, the ball was dropped onto the testing beam from three landing points in turnto obtain three waveforms. Then, the same test was carried out on the No. 2 testing beam.Finally, the waveform diagrams of the No. 1 and No. 2 testing beams at −40 ◦C, −20 ◦C,0 ◦C, 20 ◦C, 40 ◦C, and 60 ◦C were sequentially measured.
3.2. Result Analysis
According to the average value of the natural frequency obtained after three dropsof balls fall at the same temperature, the experimental results of the natural frequency atdifferent temperatures were obtained, as shown in Figure 4.
Sensors 2021, 21, x FOR PEER REVIEW 7 of 15
beam was divided into four equal lengths. The center position of the three bisectors was
marked as the landing point of small balls. Then, the acceleration sensor (Type: KT1100L,
Measurement range:5 g) was fixed with silicone rubber at a distance of 300 mm from one
end of the testing beam. The oscilloscope (Type: RIGOL MSO1104) was adjusted to the
“SINGLE” state. By controlling the air compressors (Type: AFS-550), the ball was fixed by
the fixture. The ball fell freely onto the testing beam. After landing, the height was ad-
justed until the oscilloscope showed a smooth fluctuation curve. By controlling the air
compressors, the ball was dropped onto the testing beam from three landing points in
turn to obtain three waveforms. Then, the same test was carried out on the No. 2 testing
beam. Finally, the waveform diagrams of the No. 1 and No. 2 testing beams at −40 °C, −20
°C, 0 °C, 20 °C, 40 °C, and 60 °C were sequentially measured.
3.2. Result Analysis
According to the average value of the natural frequency obtained after three drops
of balls fall at the same temperature, the experimental results of the natural frequency at
different temperatures were obtained, as shown in Figure 4.
Figure 4. Relationship between the first-grade natural frequency of the No. 1 beam and the No. 2
beam with temperature change.
Figure 4 shows the temperature-frequency fitting curve. R2 of the No. 1 beam and
No. 2 beam were both close to 1, indicating a good fit. Essentially, there was a linearly
negative correlation between the natural frequency of the testing beam and the tempera-
ture. In addition, the results of the No. 1 and No. 2 testing beams were essentially the
same. When the temperature increased by 1 °C, the fundamental frequency of the No. 1
testing beam decreased by 0.6657 Hz, with the change rate of −0.149%. When the temper-
ature increased by 1 °C, the fundamental frequency of the No. 2 testing beam decreased
by 0.6921 Hz, with the change rate of −0.154%. The above results are similar to the natural
frequency change rate of −0.16% obtained by the theoretical formula, indicating that the
experimental results were consistent with the theoretical formula.
Figure 5 shows the comparison between the theoretical and experimental results of
the first-grade natural frequency of the No. 1 and No. 2 testing beams at different
Figure 4. Relationship between the first-grade natural frequency of the No. 1 beam and the No. 2beam with temperature change.
Figure 4 shows the temperature-frequency fitting curve. R2 of the No. 1 beam and No.2 beam were both close to 1, indicating a good fit. Essentially, there was a linearly negativecorrelation between the natural frequency of the testing beam and the temperature. Inaddition, the results of the No. 1 and No. 2 testing beams were essentially the same. Whenthe temperature increased by 1 ◦C, the fundamental frequency of the No. 1 testing beamdecreased by 0.6657 Hz, with the change rate of −0.149%. When the temperature increasedby 1 ◦C, the fundamental frequency of the No. 2 testing beam decreased by 0.6921 Hz, withthe change rate of −0.154%. The above results are similar to the natural frequency changerate of −0.16% obtained by the theoretical formula, indicating that the experimental resultswere consistent with the theoretical formula.
Figure 5 shows the comparison between the theoretical and experimental results of thefirst-grade natural frequency of the No. 1 and No. 2 testing beams at different temperatures.Among them, the theoretical calculation results are derived from Formula (10).
Sensors 2021, 21, 4242 8 of 14
Sensors 2021, 21, x FOR PEER REVIEW 8 of 15
temperatures. Among them, the theoretical calculation results are derived from Formula
(10).
(a) No. 1 beam
(b) No. 2 beam
Figure 5. Relationship between theoretical results of the testing beam and experimental results of
the first-grade natural frequency with temperature change.
In Figure 5, there is little difference in the slopes between the experimental value and
the theoretical value of the fitting curve of the No. 1 and No. 2 beams, which means that
the change rate of the first-grade frequency with temperature is essentially the same.
Therefore, it can be concluded that the experimental results are consistent with the theo-
retical formula. The test value of the No. 1 beam at −40 °C had a large deviation from the
fitted curve, which may have been caused by the temperature increase of the test beam
when the test beam was taken out from the −40 °C environment into the normal tempera-
ture environment and because the experiment was not carried out immediately. Addition-
ally, the non-linearity of natural frequency below 0 °C increased with the decreasing tem-
perature, resulting in a decrease in the linear dependence of the data points overall. This
Figure 5. Relationship between theoretical results of the testing beam and experimental results of thefirst-grade natural frequency with temperature change.
In Figure 5, there is little difference in the slopes between the experimental value andthe theoretical value of the fitting curve of the No. 1 and No. 2 beams, which means that thechange rate of the first-grade frequency with temperature is essentially the same. Therefore,it can be concluded that the experimental results are consistent with the theoretical formula.The test value of the No. 1 beam at−40 ◦C had a large deviation from the fitted curve, whichmay have been caused by the temperature increase of the test beam when the test beamwas taken out from the −40 ◦C environment into the normal temperature environment andbecause the experiment was not carried out immediately. Additionally, the non-linearity ofnatural frequency below 0 ◦C increased with the decreasing temperature, resulting in adecrease in the linear dependence of the data points overall. This may have been due tothe moisture of the concrete beam. The moisture in concrete undergoes a phase change ataround 0 ◦C, which may have an impact on the natural frequency of concrete. The influence
Sensors 2021, 21, 4242 9 of 14
of water should not be omitted when the concrete is subjected to lower temperatures. Whenthe temperature was between 20 ◦C and 60 ◦C, the error between the theoretical value andthe experimental value may have been caused by the inevitable error in the experimentalprocess and the insufficient consideration of factors in the theoretical formula. Figure 5illustrates that the theoretical formula of the temperature influence coefficient is applicableto the calculation of the first-grade frequency of the simply supported beam.
3.3. Method Verification
The elastic modulus of the testing beam can be derived from the theoretical Formula (7)based on frequency and temperature. Therefore, the accuracy of the frequency-temperaturetest results can be verified by comparing the theoretical value with the elastic modulus ofthe testing beam measured by the three-point bending test.
In this paper, the three-point bending test was used to determine the elastic modulus.The fatigue testing machine was used to carry out the concentrated loading and massvalues of the No.1 and No.2 testing beams within the elastic range at a room temperatureof 20 ◦C. The test results are shown in Figure 6.
Sensors 2021, 21, x FOR PEER REVIEW 9 of 15
may have been due to the moisture of the concrete beam. The moisture in concrete under-
goes a phase change at around 0 °C, which may have an impact on the natural frequency
of concrete. The influence of water should not be omitted when the concrete is subjected
to lower temperatures. When the temperature was between 20 °C and 60 °C, the error
between the theoretical value and the experimental value may have been caused by the
inevitable error in the experimental process and the insufficient consideration of factors
in the theoretical formula. Figure 5 illustrates that the theoretical formula of the tempera-
ture influence coefficient is applicable to the calculation of the first-grade frequency of the
simply supported beam.
3.3. Method Verification
The elastic modulus of the testing beam can be derived from the theoretical Formula
(7) based on frequency and temperature. Therefore, the accuracy of the frequency-tem-
perature test results can be verified by comparing the theoretical value with the elastic
modulus of the testing beam measured by the three-point bending test.
In this paper, the three-point bending test was used to determine the elastic modulus.
The fatigue testing machine was used to carry out the concentrated loading and mass val-
ues of the No.1 and No.2 testing beams within the elastic range at a room temperature of
20 °C. The test results are shown in Figure 6.
Figure 6. Three-point bending test.
The elastic modulus of the testing beam can be obtained by the following formula:
𝐸 =𝐹
Δ𝐿∙
𝑙3
48𝐼 (12)
where,
𝐹: Concentrated loading (N);
∆𝐿: Deflection (mm);
l: Testing beam effective span length (m);
I: Section inertia moment (m4);
According to Formula (12), the test results are shown in Tables 1 and 2.
Table 1. Elastic modulus of testing beam.
Beam
Number
Initial Deflection
/mm
Initial Load
/N
Deflection
/mm
Load
/N
Elastic Modulus
/10^4 MPa
1 −0.0005 0 −0.0180 −803.36 3.0986
2 0 0 −0.0195 −910.26 3.1509
Beam
Indenter
Figure 6. Three-point bending test.
The elastic modulus of the testing beam can be obtained by the following formula:
E =F
∆L· l3
48I(12)
where,
F: Concentrated loading (N);∆L: Deflection (mm);l: Testing beam effective span length (m);I: Section inertia moment (m4);
According to Formula (12), the test results are shown in Tables 1 and 2.
Table 1. Elastic modulus of testing beam.
BeamNumber
Initial Deflection/mm
Initial Load/N
Deflection/mm
Load/N
Elastic Modulus/10ˆ4 MPa
1 −0.0005 0 −0.0180 −803.36 3.09862 0 0 −0.0195 −910.26 3.1509
Sensors 2021, 21, 4242 10 of 14
Table 2. Comparison of the elastic modulus measured by different methods (20 ◦C).
BeamNumber
Elastic Modulus/10ˆ4 MPaFrequency-Temperature Test
Calculation ValueThree Point Bending Test
Value Relative Error
1 3.1569 3.0986 1.88%2 3.1626 3.1509 0.37%
The elastic modulus calculated from the deflection value is the actual elastic modulusin the three-point bending test. The error between the calculated value and the actual valuein the frequency-temperature test of the elastic modulus of the testing beam was relativelysmall, indicating that the results of frequency-temperature test were correct.
4. Simulation Analysis on the Influence of Temperature on the Natural Frequency ofSimply Supported Beam4.1. Model Establishment
Midas/Civil is a special finite-element software for general structural analysis anddesign systems in the bridge field, which is mainly applied to the linear-elastic smalldeformation beam element calculations. Midas/Civil was used to create the testing beammodel, with size and structure consistent with the testing beam. The Y axis is the widthdirection, the X axis is the length direction, and the Z axis is the height direction. Thecalculated span length of the beam refers to the distance between two supports [27–29].Considering the distance between supports is 600 mm, the calculated span length of thebeam during the numerical simulation was set to 600 mm. The element type of the finite-element model is the beam element, in which a beam element can consist of two nodes. Themodel had a total of 30 beam elements with simply supported beam boundary conditions,which is shown in Figure 7.
Sensors 2021, 21, x FOR PEER REVIEW 10 of 15
Table 2. Comparison of the elastic modulus measured by different methods (20 °C).
Beam
Number
Elastic Modulus/10^4 MPa
Frequency-Temperature Test
Calculation Value
Three Point Bending
Test Value Relative Error
1 3.1569 3.0986 1.88%
2 3.1626 3.1509 0.37%
The elastic modulus calculated from the deflection value is the actual elastic modulus
in the three-point bending test. The error between the calculated value and the actual
value in the frequency-temperature test of the elastic modulus of the testing beam was
relatively small, indicating that the results of frequency-temperature test were correct.
4. Simulation Analysis on the Influence of Temperature on the Natural Frequency of
Simply Supported Beam
4.1. Model Establishment
Midas/Civil is a special finite-element software for general structural analysis and
design systems in the bridge field, which is mainly applied to the linear-elastic small de-
formation beam element calculations. Midas/Civil was used to create the testing beam
model, with size and structure consistent with the testing beam. The Y axis is the width
direction, the X axis is the length direction, and the Z axis is the height direction. The
calculated span length of the beam refers to the distance between two supports [27–29].
Considering the distance between supports is 600 mm, the calculated span length of the
beam during the numerical simulation was set to 600 mm. The element type of the finite-
element model is the beam element, in which a beam element can consist of two nodes.
The model had a total of 30 beam elements with simply supported beam boundary con-
ditions, which is shown in Figure 7.
Figure 7. The testing beam model.
In order to make the simulation results more accurate, the actual values of the elastic
modulus at 20 °C obtained from the three-point bending test were substituted into the
design parameters of the testing beam model and verified with Formula (1).
4.2. Result Analysis
The first four-grades natural frequency of the testing beam model obtained at a fun-
damental frequency temperature of 20 °C were taken. Then, the first four-grades natural
frequency of the testing beam model at −40 °C, −20 °C, 0 °C, 20 °C, 40 °C, and 60 °C was
successively obtained. The relationship between n-grades natural frequency (n = 1, 2, 3, 4)
and temperature in the theoretical results and simulation results of beam 1 and beam 2 is
shown in Figure 8.
Figure 7. The testing beam model.
In order to make the simulation results more accurate, the actual values of the elasticmodulus at 20 ◦C obtained from the three-point bending test were substituted into thedesign parameters of the testing beam model and verified with Formula (1).
4.2. Result Analysis
The first four-grades natural frequency of the testing beam model obtained at afundamental frequency temperature of 20 ◦C were taken. Then, the first four-gradesnatural frequency of the testing beam model at −40 ◦C, −20 ◦C, 0 ◦C, 20 ◦C, 40 ◦C, and60 ◦C was successively obtained. The relationship between n-grades natural frequency(n = 1, 2, 3, 4) and temperature in the theoretical results and simulation results of beam 1and beam 2 is shown in Figure 8.
Sensors 2021, 21, 4242 11 of 14Sensors 2021, 21, x FOR PEER REVIEW 11 of 15
(a) 1st frequency of beam 1 (b) 1st frequency of beam 2
(c) 2st frequency of beam 1 (d) 2st frequency of beam 2
(e) 3st frequency of beam 1 (f) 3st frequency of beam 2
Figure 8. Cont.
Sensors 2021, 21, 4242 12 of 14Sensors 2021, 21, x FOR PEER REVIEW 12 of 15
(g) 4st frequency of beam 1 (h) 4st frequency of beam 2
Figure 8. Relationship between the n-grades natural frequency (n = 1, 2, 3, 4) and temperature of the
testing beam’s theoretical and simulation results.
According to Figure 8a,b, the residual square and R2 of temperature-frequency fitting
curve of the No. 1 beam and No. 2 beam were both close to one, indicating that fitting was
good. Both the theoretical and simulation results showed that the first four-grades modal
frequency of the testing beam were linearly negatively correlated with temperature. Ac-
cording to the frequency-temperature fitting curve, the theoretical and simulated values
of the first-grade natural frequency of the No. 1 beam and No. 2 beam had little difference
within the temperature range of −40 °C to 60 °C, which were basically consistent. In terms
of the change rate, when the temperature of the No.1 beam increased by 1 °C, the theoret-
ical value decreased by 0.148%, and the simulated value decreased by 0.149%. The results
were essentially the same for both. When the temperature of the No. 2 beam increased by
1 °C, the theoretical value decreased by 0.148%, and the simulated value decreased by
0.150%. The results of both are also almost the same.
Figure 8c,d shows the relationship between the theoretical and simulated second-
order natural frequency and temperature of the No. 1 beam and No. 2 beam, respectively.
According to the frequency-temperature fitting curve, it was found that the theoretical
value of the second-order natural frequency was greater than the simulated value in the
temperature range of −40~60 °C. At 20 °C, the theoretical value of the No. 1 beam was
1777.306 Hz, and the simulated value was 1412.862 Hz. Under the same temperature con-
ditions, the theoretical value of the No.2 beam was 1792.243 Hz, and the simulated value
was 1424.735 Hz. Based on the theoretical value, the simulated value was about 20%
smaller than the theoretical value, indicating that there was a relatively large error. In
terms of the change rate, for every 1 °C increase in temperature, the theoretical value de-
creased by 0.148% and the simulated value decreased by 0.148%. That is to say, the two
were basically consistent in terms of the change trend.
Figures 8e–h show the relationship between theoretical and simulated third-order
and fourth-order natural frequencies and temperature of the No. 1 beam and No. 2 beam.
From the frequency-temperature fitting curve, it was found that the variation trend of the
third-order natural frequency and fourth-order natural frequency of the No. 1 beam was
similar to that of the second-order natural frequency in the temperature range of −40~60
°C. In addition, the slope of the curve increased with the increase of order. However, the
theoretical increase in the slope of the curve was much greater than the simulated one. At
20 °C, the theoretical value of the third-order natural frequency of the No.1 beam was
3998.94 Hz, and the simulated value was 2335.15 Hz. In contrast, the simulated value was
41.61% lower than the theoretical value. At the same temperature, the theoretical value of
the fourth-order natural frequency was 7109.23 Hz, and the simulated value was 2818.29
Figure 8. Relationship between the n-grades natural frequency (n = 1, 2, 3, 4) and temperature of thetesting beam’s theoretical and simulation results.
According to Figure 8a,b, the residual square and R2 of temperature-frequency fittingcurve of the No. 1 beam and No. 2 beam were both close to one, indicating that fittingwas good. Both the theoretical and simulation results showed that the first four-gradesmodal frequency of the testing beam were linearly negatively correlated with temperature.According to the frequency-temperature fitting curve, the theoretical and simulated valuesof the first-grade natural frequency of the No. 1 beam and No. 2 beam had little differencewithin the temperature range of −40 ◦C to 60 ◦C, which were basically consistent. In termsof the change rate, when the temperature of the No.1 beam increased by 1 ◦C, the theoreticalvalue decreased by 0.148%, and the simulated value decreased by 0.149%. The results wereessentially the same for both. When the temperature of the No. 2 beam increased by 1 ◦C,the theoretical value decreased by 0.148%, and the simulated value decreased by 0.150%.The results of both are also almost the same.
Figure 8c,d shows the relationship between the theoretical and simulated second-order natural frequency and temperature of the No. 1 beam and No. 2 beam, respectively.According to the frequency-temperature fitting curve, it was found that the theoreticalvalue of the second-order natural frequency was greater than the simulated value in thetemperature range of −40~60 ◦C. At 20 ◦C, the theoretical value of the No. 1 beam was1777.306 Hz, and the simulated value was 1412.862 Hz. Under the same temperatureconditions, the theoretical value of the No.2 beam was 1792.243 Hz, and the simulatedvalue was 1424.735 Hz. Based on the theoretical value, the simulated value was about20% smaller than the theoretical value, indicating that there was a relatively large error.In terms of the change rate, for every 1 ◦C increase in temperature, the theoretical valuedecreased by 0.148% and the simulated value decreased by 0.148%. That is to say, the twowere basically consistent in terms of the change trend.
Figure 8e–h show the relationship between theoretical and simulated third-order andfourth-order natural frequencies and temperature of the No. 1 beam and No. 2 beam. Fromthe frequency-temperature fitting curve, it was found that the variation trend of the third-order natural frequency and fourth-order natural frequency of the No. 1 beam was similarto that of the second-order natural frequency in the temperature range of −40~60 ◦C.In addition, the slope of the curve increased with the increase of order. However, thetheoretical increase in the slope of the curve was much greater than the simulated one.At 20 ◦C, the theoretical value of the third-order natural frequency of the No.1 beam was3998.94 Hz, and the simulated value was 2335.15 Hz. In contrast, the simulated valuewas 41.61% lower than the theoretical value. At the same temperature, the theoreticalvalue of the fourth-order natural frequency was 7109.23 Hz, and the simulated value was2818.29 Hz. In contrast, the simulated value was 60.36% lower than the theoretical value.Furthermore, the comparison result of the No. 2 beam was also similar. With the theoretical
Sensors 2021, 21, 4242 13 of 14
value as the standard, the high-order natural frequency of the beam will cause a largeerror. According to the theoretical Formula (7), it was found that the n-order frequencyof the simply supported beam was n2 times the first-order frequency. Although the first-order frequency obtained by theoretical calculation was close to the simulated first-orderfrequency, there were still errors. With the increase of the order, the error between thetheoretical frequency and the simulated frequency will further increase. However, thesimulation results had the same trend as the theoretical results. In other words, therewas a linear relationship between frequency and temperature change. The simulationresults showed that the theoretical calculation formula is only applicable for calculatingthe first-grade frequency of the simply supported beam, not for the frequencies above thefirst grade, which provides references for the study on natural vibration characteristics ofactual bridges.
5. Conclusions
This paper obtained the relationship between the natural frequency of simply sup-ported beams and temperature through theoretical analysis, experimental testing, andnumerical simulation. The conclusions are as follows:
(1) The theoretical calculation formula for the change of natural frequency with tempera-ture was obtained through the theoretical analysis of natural vibration characteristicsof simply supported reinforced concrete beams. Theoretical analysis indicated thatthe effect of temperature on the natural frequency of the simply supported beamwas mainly caused by the change of the elastic modulus of simply supported beammaterial. There was little influence of structure size on natural frequency.
(2) The influence of temperature in the range of −40~60 ◦C on the natural frequencyof the simply supported beam was analyzed through the small ball excitation test.The temperature range was so wide that it covered the conventional service tempera-ture range of simply supported beam bridges. An acceleration sensor was used tomonitor the natural frequency of the simply supported beam. The natural frequency-temperature change curve was obtained and verified. The curve showed that, for each1 ◦C increase in temperature in the range of −40 ◦C to 60 ◦C, the natural frequency ofthe simply supported beam decreased by 0.148%.
(3) The Midas/Civil finite-element simulation was applied to analyze the natural fre-quency of simply supported beams at different temperatures. By comparing with thetheoretical values, the theoretical calculation formula for the natural frequency of thesimply supported beam was only applicable to the fundamental frequency, ratherthan the frequencies above the first grade.
The evolution law of the natural frequency of reinforced concrete simply supportedbeams with temperature changes can provide references for dynamic characteristics moni-toring and damage identification of simply supported beam bridges. The actual bridgestructure and the influence of uneven temperature distribution on the bridge’s naturalvibration characteristics need to be considered in subsequent studies.
Author Contributions: Methodology, K.Z., C.L. and L.W.; investigation, Y.C. and Z.Y.; formalanalysis, Z.Y. and C.L.; writing—original draft preparation, Y.C., K.Z., and C.L.; writing—reviewand editing, Y.C., K.L. and L.W.; supervision, K.L. and L.W. All authors have read and agreed to thepublished version of the manuscript.
Funding: This research was funded by Beijing major science and technology projects, grant numberZ191100008019002, Fundamental Research Funds for the Central University (FRF-TP-19-050A1,FRF-BD-19-001A, FRF-MP-19-014).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data is contained within the article.
Sensors 2021, 21, 4242 14 of 14
Conflicts of Interest: The authors declare no conflict of interest.
References1. Zong, Z.; Ren, W.; Zhen, Z. Load-carrying capacity assessment methods of existing bridges. Earthq. Eng. Eng. Dyn. 2005, 25,
147–152.2. Gillich, G.-R.; Furdui, H.; Wahab, M.A.; Korka, Z.-I. A robust damage detection method based on multi-modal analysis in variable
temperature conditions. Mech. Syst. Signal Process. 2019, 115, 361–379. [CrossRef]3. Hu, W.-H.; Tang, D.-H.; Teng, J.; Said, S.; Rohrmann, R.G. Structural Health Monitoring of a Prestressed Concrete Bridge Based on
Statistical Pattern Recognition of Continuous Dynamic Measurements over 14 years. Sensors 2018, 18, 4117. [CrossRef] [PubMed]4. Wang, X.; Gao, Q.; Liu, Y. Damage Detection of Bridges under Environmental Temperature Changes Using a Hybrid Method.
Sensors 2020, 20, 3999. [CrossRef] [PubMed]5. Liu, H.; Wang, X.; Jiao, Y. Effect of Temperature Variation on Modal Frequency of Reinforced Concrete Slab and Beam in Cold
Regions. Shock. Vib. 2016, 2016, 1–17. [CrossRef]6. Shan, W.; Wang, X.; Jiao, Y. Modeling of Temperature Effect on Modal Frequency of Concrete Beam Based on Field Monitoring
Data. Shock. Vib. 2018, 2018, 1–13. [CrossRef]7. Jang, J.; Smyth, A.W. Data-driven models for temperature distribution effects on natural frequencies and thermal prestress
modeling. Struct. Control Health Monit. 2020, 27, 2489. [CrossRef]8. Teng, J.; Tang, D.-H.; Zhang, X.; Hu, W.-H.; Said, S.; Rohrmann, R.G. Automated Modal Analysis for Tracking Structural Change
during Construction and Operation Phases. Sensors 2019, 19, 927. [CrossRef]9. Liu, J.; Tang, K. Advances of Vibration-based Damage Identification Methods. Acta Sci. Nat. Univ. Sunyatseni 2004, 43, 57–61.10. Farrar, C.R.; Doebling, S.W.; Cornwell, P.J.; Straser, E.G. Variability of Modal Parameters Measured on the Alamosa Canyon Bridge;
USDOE Assistant Secretary for Human Resources and Administration: Washington, DC, USA, 1997; pp. 257–263.11. Liu, C.; DeWolf, J.T. Effect of Temperature on Modal Variability of a Curved Concrete Bridge under Ambient Loads. J. Struct. Eng.
2007, 133, 1742–1751. [CrossRef]12. Sun, L.; Zhou, Y.; Min, Z. Experimental Study on the Effect of Temperature on Modal Frequencies of Bridges. Int. J. Struct. Stab.
Dyn. 2018, 18, 1850155. [CrossRef]13. Zhou, Y.; Sun, L. Effects of environmental and operational actions on the modal frequency variations of a sea-crossing bridge: A
periodicity perspective. Mech. Syst. Signal Process. 2019, 131, 505–523. [CrossRef]14. Peeters, B.; De Roeck, G. One-year monitoring of the Z24-Bridge: Environmental effects versus damage events. Earthq. Eng.
Struct. Dyn. 2001, 30, 149–171. [CrossRef]15. Xia, Y.; Hao, H.; Zanardo, G.; Deeks, A. Long term vibration monitoring of an RC slab: Temperature and humidity effect. Eng.
Struct. 2006, 28, 441–452. [CrossRef]16. Kim, J.-T.; Park, J.-H.; Lee, B.-J. Vibration-based damage monitoring in model plate-girder bridges under uncertain temperature
conditions. Eng. Struct. 2007, 29, 1354–1365. [CrossRef]17. Hu, T. The Influence of Temperature on the Self Vibration Characteristics of Cfst Deck Arch Bridge and Seismic Response Analysis; Southwest
Jiaotong University: Chengdu, China, 2018.18. Teng, J.; Tang, D.-H.; Hu, W.-H.; Lu, W.; Feng, Z.-W.; Ao, C.-F.; Liao, M.-H. Mechanism of the effect of temperature on frequency
based on long-term monitoring of an arch bridge. Struct. Health Monit. 2020. [CrossRef]19. Cornwell, P.; Farrar, C.; Doebling, S.; Sohn, H. Environmental Variability of Modal Properties. Exp. Tech. 1999, 23, 45–48.
[CrossRef]20. Xu, Y.; Zhu, S.; Zong, Z. Experimental study on effects of environmental temperature on dynamic characteristics of bridge
structures. Earthq. Eng. Eng. Dyn. 2007, 27, 119–123.21. Zhang, J. Regression analysis of environmental temperature and humidity and the natural frequencies of composite beam. J.
Hebei Univ. Eng. (Nat. Sci. Ed.) 2012, 29, 16–19.22. Li, X.; Chen, A.; Ma, R. Influence of temperature on modal parameters of bridge. J. South China Univ. Technol. (Nat. Sci. Ed.) 2012,
40, 138–143.23. Wang, J. Research on Natural Frequency of Simply Supported Beam Basis on Multiple Temperature Fields; Jilin University: Changchun,
China, 2012.24. Yang, M. Evolvement of Nature Vibration Characteriscs of Concrete Girder Bridge Considering Environmental Temperature and Time-
Depended Performance of Concrete; Beijing Jiaotong University: Beijing, China, 2014.25. Li, Y. Influence of the Temperature to the Evolvement of Dynamic Characteriscs of Concrete Girder Bridge; Beijing Jiaotong University:
Beijing, China, 2016.26. CEB-FIP. Model Code 2010. In Fib Special Activity Group 5; The International Federation for Structural Concrete: Lausanne,
Switzerland, 2010.27. CCCC Highway Consultants Co., Ltd. General Specifications for Design of Highway Bridges and Culverts, 1st ed.; China Communica-
tions Press: Beijing, China, 2015; ISBN 978-7-114-12506-5.28. Sharma, G.S.; Sarkar, A. Directivity based control of acoustic radiation. Appl. Acoust. 2019, 154, 226–235. [CrossRef]29. Sharma, G.S.; Sarkar, A. Directivity-Based Passive Barrier for Local Control of Low-Frequency Noise. J. Comput. Acoust. 2017, 26,
1850012. [CrossRef]